@lumifai/harness-tool-pack-file-processing 0.0.1

package/README.md

@@ -0,0 +1,13 @@
+# @lumifai/harness-tool-pack-file-processing
+
+Extract text from workspace PDF files and write results under `/processed/`.
+
+```ts
+import { createFileProcessingToolPack } from '@lumifai/harness-tool-pack-file-processing';
+
+createFileProcessingToolPack();
+```
+
+`process_pdf` reads a workspace PDF, runs OCR-aware extraction via LiteParse, and returns an `output_path` reference instead of inline content. Bundled agent skills ship in this package and are merged into `workspace.skills` when the pack is enabled.
+
+Tool pack id: `file-processing`. Disabled by default (`enabledByDefault: false`). Default policy: `process_pdf` → `ask`.

package/dist/file-processing.d.ts

@@ -0,0 +1,3 @@
+import type { ToolsInput } from '@mastra/core/agent';
+export declare function createFileProcessingTools(): ToolsInput;
+export declare function createFileProcessingToolPack(): import("@lumifai/harness-tool-packs").ToolPack;

package/dist/file-processing.js

@@ -0,0 +1,86 @@
+"use strict";
+var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    var desc = Object.getOwnPropertyDescriptor(m, k);
+    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
+      desc = { enumerable: true, get: function() { return m[k]; } };
+    }
+    Object.defineProperty(o, k2, desc);
+}) : (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    o[k2] = m[k];
+}));
+var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
+    Object.defineProperty(o, "default", { enumerable: true, value: v });
+}) : function(o, v) {
+    o["default"] = v;
+});
+var __importStar = (this && this.__importStar) || (function () {
+    var ownKeys = function(o) {
+        ownKeys = Object.getOwnPropertyNames || function (o) {
+            var ar = [];
+            for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
+            return ar;
+        };
+        return ownKeys(o);
+    };
+    return function (mod) {
+        if (mod && mod.__esModule) return mod;
+        var result = {};
+        if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
+        __setModuleDefault(result, mod);
+        return result;
+    };
+})();
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.createFileProcessingTools = createFileProcessingTools;
+exports.createFileProcessingToolPack = createFileProcessingToolPack;
+const tools_1 = require("@mastra/core/tools");
+const zod_1 = require("zod");
+const harness_tool_packs_1 = require("@lumifai/harness-tool-packs");
+const utils_1 = require("./utils");
+const workspaceFilePathSchema = zod_1.z
+    .string()
+    .min(1)
+    .describe('Absolute workspace path to a file, e.g. /documents/report.pdf');
+function createFileProcessingTools() {
+    const processPdf = (0, tools_1.createTool)({
+        id: 'process_pdf',
+        description: 'Extract text content from a PDF file and write it to /processed/ as plain text. Returns a reference path, not the content itself.',
+        inputSchema: zod_1.z.object({
+            path: workspaceFilePathSchema,
+        }),
+        execute: async (inputData, context) => {
+            const { LiteParse } = await Promise.resolve().then(() => __importStar(require('@llamaindex/liteparse')));
+            const { normalizedPath, buffer } = await (0, utils_1.readWorkspaceBinaryFile)(context, inputData.path);
+            const parser = new LiteParse({ ocrEnabled: true, quiet: true });
+            const result = await parser.parse(buffer);
+            const outputContent = result.pages
+                .map((page) => `--- Page ${page.pageNum} ---\n${page.text}`)
+                .join('\n\n');
+            const outputPath = (0, utils_1.deriveOutputPath)(normalizedPath, 'txt');
+            await (0, utils_1.writeProcessedText)(context, outputPath, outputContent);
+            return {
+                source_path: normalizedPath,
+                output_path: outputPath,
+                page_count: result.pages.length,
+                char_count: outputContent.length,
+            };
+        },
+    });
+    return {
+        processPdf,
+    };
+}
+function createFileProcessingToolPack() {
+    return (0, harness_tool_packs_1.defineToolPack)({
+        id: 'file-processing',
+        description: 'Extract text from PDF files.',
+        enabledByDefault: false,
+        tools: createFileProcessingTools(),
+        permissionPolicies: {
+            process_pdf: 'ask',
+        },
+    });
+}
+//# sourceMappingURL=file-processing.js.map

package/dist/file-processing.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"file-processing.js","sourceRoot":"","sources":["../src/file-processing.ts"],"names":[],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAaA,8DAkCC;AAED,oEAUC;AA1DD,8CAAgD;AAChD,6BAAwB;AAExB,oEAA6D;AAE7D,mCAAwF;AAExF,MAAM,uBAAuB,GAAG,OAAC;KAC9B,MAAM,EAAE;KACR,GAAG,CAAC,CAAC,CAAC;KACN,QAAQ,CAAC,+DAA+D,CAAC,CAAC;AAE7E,SAAgB,yBAAyB;IACvC,MAAM,UAAU,GAAG,IAAA,kBAAU,EAAC;QAC5B,EAAE,EAAE,aAAa;QACjB,WAAW,EACT,mIAAmI;QACrI,WAAW,EAAE,OAAC,CAAC,MAAM,CAAC;YACpB,IAAI,EAAE,uBAAuB;SAC9B,CAAC;QACF,OAAO,EAAE,KAAK,EAAE,SAAS,EAAE,OAAO,EAAE,EAAE;YACpC,MAAM,EAAE,SAAS,EAAE,GAAG,wDAAa,uBAAuB,GAAC,CAAC;YAC5D,MAAM,EAAE,cAAc,EAAE,MAAM,EAAE,GAAG,MAAM,IAAA,+BAAuB,EAAC,OAAO,EAAE,SAAS,CAAC,IAAI,CAAC,CAAC;YAE1F,MAAM,MAAM,GAAG,IAAI,SAAS,CAAC,EAAE,UAAU,EAAE,IAAI,EAAE,KAAK,EAAE,IAAI,EAAE,CAAC,CAAC;YAChE,MAAM,MAAM,GAAG,MAAM,MAAM,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC;YAE1C,MAAM,aAAa,GAAG,MAAM,CAAC,KAAK;iBAC/B,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,YAAY,IAAI,CAAC,OAAO,SAAS,IAAI,CAAC,IAAI,EAAE,CAAC;iBAC3D,IAAI,CAAC,MAAM,CAAC,CAAC;YAEhB,MAAM,UAAU,GAAG,IAAA,wBAAgB,EAAC,cAAc,EAAE,KAAK,CAAC,CAAC;YAC3D,MAAM,IAAA,0BAAkB,EAAC,OAAO,EAAE,UAAU,EAAE,aAAa,CAAC,CAAC;YAE7D,OAAO;gBACL,WAAW,EAAE,cAAc;gBAC3B,WAAW,EAAE,UAAU;gBACvB,UAAU,EAAE,MAAM,CAAC,KAAK,CAAC,MAAM;gBAC/B,UAAU,EAAE,aAAa,CAAC,MAAM;aACjC,CAAC;QACJ,CAAC;KACF,CAAC,CAAC;IAEH,OAAO;QACL,UAAU;KACX,CAAC;AACJ,CAAC;AAED,SAAgB,4BAA4B;IAC1C,OAAO,IAAA,mCAAc,EAAC;QACpB,EAAE,EAAE,iBAAiB;QACrB,WAAW,EAAE,8BAA8B;QAC3C,gBAAgB,EAAE,KAAK;QACvB,KAAK,EAAE,yBAAyB,EAAE;QAClC,kBAAkB,EAAE;YAClB,WAAW,EAAE,KAAK;SACnB;KACF,CAAC,CAAC;AACL,CAAC"}

package/dist/index.d.ts

@@ -0,0 +1 @@
+export { createFileProcessingTools, createFileProcessingToolPack } from './file-processing';

package/dist/index.js

@@ -0,0 +1,7 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.createFileProcessingToolPack = exports.createFileProcessingTools = void 0;
+var file_processing_1 = require("./file-processing");
+Object.defineProperty(exports, "createFileProcessingTools", { enumerable: true, get: function () { return file_processing_1.createFileProcessingTools; } });
+Object.defineProperty(exports, "createFileProcessingToolPack", { enumerable: true, get: function () { return file_processing_1.createFileProcessingToolPack; } });
+//# sourceMappingURL=index.js.map

package/dist/index.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.js","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":";;;AAAA,qDAA4F;AAAnF,4HAAA,yBAAyB,OAAA;AAAE,+HAAA,4BAA4B,OAAA"}

package/dist/utils.d.ts

@@ -0,0 +1,8 @@
+import type { ToolExecutionContext } from '@mastra/core/tools';
+export declare function normalizeWorkspaceFilePath(path: string): string;
+export declare function readWorkspaceBinaryFile(context: ToolExecutionContext, path: string): Promise<{
+    normalizedPath: string;
+    buffer: Buffer;
+}>;
+export declare function writeProcessedText(context: ToolExecutionContext, outputPath: string, content: string): Promise<void>;
+export declare function deriveOutputPath(sourcePath: string, suffix: string): string;

package/dist/utils.js

@@ -0,0 +1,41 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.normalizeWorkspaceFilePath = normalizeWorkspaceFilePath;
+exports.readWorkspaceBinaryFile = readWorkspaceBinaryFile;
+exports.writeProcessedText = writeProcessedText;
+exports.deriveOutputPath = deriveOutputPath;
+function normalizeWorkspaceFilePath(path) {
+    const trimmed = path.trim();
+    if (!trimmed.startsWith('/')) {
+        throw new Error(`Path must be absolute within the workspace: ${path}`);
+    }
+    return trimmed;
+}
+async function readWorkspaceBinaryFile(context, path) {
+    const normalizedPath = normalizeWorkspaceFilePath(path);
+    const workspace = context.workspace;
+    if (!workspace?.filesystem?.readFile) {
+        throw new Error(`File processing tools require a workspace filesystem to read ${normalizedPath}`);
+    }
+    const fileContent = await workspace.filesystem.readFile(normalizedPath);
+    const buffer = Buffer.isBuffer(fileContent) ? fileContent : Buffer.from(fileContent);
+    return { normalizedPath, buffer };
+}
+async function writeProcessedText(context, outputPath, content) {
+    const workspace = context.workspace;
+    if (!workspace?.filesystem?.writeFile) {
+        throw new Error(`File processing tools require a workspace filesystem to write ${outputPath}`);
+    }
+    const dir = outputPath.substring(0, outputPath.lastIndexOf('/'));
+    if (dir && dir !== '/') {
+        await workspace.filesystem.mkdir(dir, { recursive: true });
+    }
+    await workspace.filesystem.writeFile(outputPath, content);
+}
+function deriveOutputPath(sourcePath, suffix) {
+    const parts = sourcePath.split('/');
+    const filename = parts[parts.length - 1] ?? sourcePath;
+    parts[parts.length - 1] = `${filename}.${suffix}`;
+    return parts.join('/');
+}
+//# sourceMappingURL=utils.js.map

package/dist/utils.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"utils.js","sourceRoot":"","sources":["../src/utils.ts"],"names":[],"mappings":";;AAEA,gEAMC;AAED,0DAeC;AAED,gDAgBC;AAED,4CAKC;AAhDD,SAAgB,0BAA0B,CAAC,IAAY;IACrD,MAAM,OAAO,GAAG,IAAI,CAAC,IAAI,EAAE,CAAC;IAC5B,IAAI,CAAC,OAAO,CAAC,UAAU,CAAC,GAAG,CAAC,EAAE,CAAC;QAC7B,MAAM,IAAI,KAAK,CAAC,+CAA+C,IAAI,EAAE,CAAC,CAAC;IACzE,CAAC;IACD,OAAO,OAAO,CAAC;AACjB,CAAC;AAEM,KAAK,UAAU,uBAAuB,CAC3C,OAA6B,EAC7B,IAAY;IAEZ,MAAM,cAAc,GAAG,0BAA0B,CAAC,IAAI,CAAC,CAAC;IACxD,MAAM,SAAS,GAAG,OAAO,CAAC,SAAS,CAAC;IACpC,IAAI,CAAC,SAAS,EAAE,UAAU,EAAE,QAAQ,EAAE,CAAC;QACrC,MAAM,IAAI,KAAK,CACb,gEAAgE,cAAc,EAAE,CACjF,CAAC;IACJ,CAAC;IAED,MAAM,WAAW,GAAG,MAAM,SAAS,CAAC,UAAU,CAAC,QAAQ,CAAC,cAAc,CAAC,CAAC;IACxE,MAAM,MAAM,GAAG,MAAM,CAAC,QAAQ,CAAC,WAAW,CAAC,CAAC,CAAC,CAAC,WAAW,CAAC,CAAC,CAAC,MAAM,CAAC,IAAI,CAAC,WAAW,CAAC,CAAC;IACrF,OAAO,EAAE,cAAc,EAAE,MAAM,EAAE,CAAC;AACpC,CAAC;AAEM,KAAK,UAAU,kBAAkB,CACtC,OAA6B,EAC7B,UAAkB,EAClB,OAAe;IAEf,MAAM,SAAS,GAAG,OAAO,CAAC,SAAS,CAAC;IACpC,IAAI,CAAC,SAAS,EAAE,UAAU,EAAE,SAAS,EAAE,CAAC;QACtC,MAAM,IAAI,KAAK,CAAC,iEAAiE,UAAU,EAAE,CAAC,CAAC;IACjG,CAAC;IAED,MAAM,GAAG,GAAG,UAAU,CAAC,SAAS,CAAC,CAAC,EAAE,UAAU,CAAC,WAAW,CAAC,GAAG,CAAC,CAAC,CAAC;IACjE,IAAI,GAAG,IAAI,GAAG,KAAK,GAAG,EAAE,CAAC;QACvB,MAAM,SAAS,CAAC,UAAU,CAAC,KAAK,CAAC,GAAG,EAAE,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAC;IAC7D,CAAC;IAED,MAAM,SAAS,CAAC,UAAU,CAAC,SAAS,CAAC,UAAU,EAAE,OAAO,CAAC,CAAC;AAC5D,CAAC;AAED,SAAgB,gBAAgB,CAAC,UAAkB,EAAE,MAAc;IACjE,MAAM,KAAK,GAAG,UAAU,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC;IACpC,MAAM,QAAQ,GAAG,KAAK,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC,CAAC,IAAI,UAAU,CAAC;IACvD,KAAK,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC,CAAC,GAAG,GAAG,QAAQ,IAAI,MAAM,EAAE,CAAC;IAClD,OAAO,KAAK,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC;AACzB,CAAC"}

package/package.json

@@ -0,0 +1,37 @@
+{
+  "name": "@lumifai/harness-tool-pack-file-processing",
+  "version": "0.0.1",
+  "license": "UNLICENSED",
+  "engines": {
+    "node": ">=22.13.0"
+  },
+  "exports": {
+    ".": {
+      "types": "./dist/index.d.ts",
+      "import": "./dist/index.js",
+      "require": "./dist/index.js",
+      "default": "./dist/index.js"
+    },
+    "./package.json": "./package.json"
+  },
+  "main": "dist/index.js",
+  "types": "dist/index.d.ts",
+  "files": [
+    "dist",
+    "!dist/**/*.test.*",
+    "skills"
+  ],
+  "dependencies": {
+    "@llamaindex/liteparse": "^2.0.8",
+    "@mastra/core": "1.50.1",
+    "zod": "^4.4.3",
+    "@lumifai/harness-tool-packs": "0.0.1"
+  },
+  "publishConfig": {
+    "access": "public",
+    "registry": "https://registry.npmjs.org/"
+  },
+  "scripts": {
+    "build": "tsc -p tsconfig.json"
+  }
+}

package/skills/file-processing/SKILL.md

@@ -0,0 +1,29 @@
+---
+name: file-processing
+description: Extract text from PDF files for LLM consumption
+version: 1.1.0
+tags:
+  - file-processing
+  - pdf
+---
+
+# File Processing Tools
+
+Use the file-processing tool pack to extract text from PDF documents. Extracted content is written to the workspace filesystem as plain text files, keeping your context window clean.
+
+## Available Tools
+
+| Tool          | Format | Output                 |
+| ------------- | ------ | ---------------------- |
+| `process_pdf` | PDF    | Text with page markers |
+
+## Workflow
+
+1. Call `process_pdf` with the absolute workspace path to the file.
+2. The tool writes extracted text to `/processed/<filename>.txt`.
+3. Use the returned `output_path` to reference the extracted content in subsequent operations.
+
+## Tips
+
+- All tools return metadata only (paths, counts) — the extracted text stays in the filesystem.
+- Processed files are written to `/processed/` to keep them organized and separate from source files.